node-prompts

What it teaches: a pure-Node Tesseron integration whose domain revolves around ctx.sample and ctx.elicit. No Express, no HTTP server - just a Node script that registers actions and connects. Good when you're building a CLI, a daemon, or a worker that Claude should drive, and the agent's LLM is part of the workflow.

Source: examples/node-prompts

Run it

pnpm --filter node-prompts dev
# prints the claim code to stdout; no browser

Domain

A library of reusable LLM prompts. Claude can:

addPrompt, listPrompts, deletePrompt, importPrompts, purgeAll - CRUD over the library.
testPrompt - fill {{var}} placeholders, send the prompt through ctx.sample, store the response as lastTest.
refinePrompt - elicit a free-text refinement instruction from the user via ctx.elicit, then ask the agent LLM via ctx.sample to rewrite the template in place.
generateVariants - ask the agent LLM for N alternative phrasings of a prompt, stream progress as each lands, store them as new prompts.

Two subscribable resources push updates on every mutation: library (Prompt[]) and lastTest (TestResult | null).

What's inside

import { tesseron } from '@tesseron/server';
import { z } from 'zod';

tesseron.app({ id: 'prompt_lab', name: 'Prompt Lab' });

tesseron.action('testPrompt')
  .input(z.object({
    id: z.string(),
    variables: z.record(z.string(), z.string()).optional(),
  }))
  .handler(async ({ id, variables }, ctx) => {
    const prompt = prompts.get(id)!;
    if (!ctx.agentCapabilities.sampling) {
      throw new Error('Agent does not support sampling.');
    }
    const filled = applyTemplate(prompt.template, variables ?? {});
    const response = await ctx.sample({ prompt: filled, maxTokens: 512 });
    // store response as lastTest, bump timesTested, notify subscribers
    return { id, response };
  });

tesseron.action('refinePrompt')
  .input(z.object({ id: z.string() }))
  .handler(async ({ id }, ctx) => {
    const answer = await ctx.elicit({
      question: `What should change?`,
      schema: z.object({ instruction: z.string().min(1) }),
      jsonSchema: { /* ... */ },
    });
    if (answer === null) return { id, refined: false, cancelled: true };
    const rewritten = await ctx.sample({
      prompt: `Rewrite this prompt per instruction: ${answer.instruction}\n\n${prompt.template}`,
    });
    // replace template with rewritten.trim(), notify subscribers
  });

tesseron.resource<Prompt[]>('library')
  .read(() => Array.from(prompts.values()))
  .subscribe((emit) => { librarySubs.add(emit); return () => librarySubs.delete(emit); });

const welcome = await tesseron.connect();
log(`Tesseron ready. Claim code: ${welcome.claimCode}`);

Features exercised: actions, annotations, subscribable resources, ctx.confirm (deletePrompt), ctx.elicit with schema (refinePrompt, purgeAll), ctx.progress (importPrompts, generateVariants), ctx.sample free-text (testPrompt, refinePrompt) and ctx.sample with Zod schema (generateVariants), cancellation via ctx.signal, capability gating via ctx.agentCapabilities.sampling, structured logging via log(), signal-aware shutdown.

Pair with express-prompts to see the same domain served over HTTP.